This article was downloaded by: [University of Connecticut] On: 11 October 2014, At: 08:01 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
The Journal of Genetic Psychology: Research and Theory on Human Development Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/vgnt20
Information Transmission in Young Children: When Social Information Is More Important Than Nonsocial Information a
Nicola McGuigan & Marcus Cubillo
a
a
Heriot-Watt University , Scotland Accepted author version posted online: 24 Apr 2013.Published online: 02 Oct 2013.
To cite this article: Nicola McGuigan & Marcus Cubillo (2013) Information Transmission in Young Children: When Social Information Is More Important Than Nonsocial Information, The Journal of Genetic Psychology: Research and Theory on Human Development, 174:6, 605-619, DOI: 10.1080/00221325.2012.749833 To link to this article: http://dx.doi.org/10.1080/00221325.2012.749833
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages,
and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.
Downloaded by [University of Connecticut] at 08:01 11 October 2014
This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions
Downloaded by [University of Connecticut] at 08:01 11 October 2014
The Journal of Genetic Psychology, 2013, 174(6), 605–619 C Taylor & Francis Group, LLC Copyright doi: 10.1080/00221325.2012.749833
Information Transmission in Young Children: When Social Information Is More Important Than Nonsocial Information NICOLA McGUIGAN MARCUS CUBILLO Heriot-Watt University, Scotland
ABSTRACT. The authors’ aim was to use a highly novel open diffusion paradigm to investigate the transmission of social information (i.e., gossip) and general knowledge within 2 groups of 10- and 11-year-old children. Four children, 2 from each group, acted as a primed information source, selected on the basis of sex and dominance ranking (high or low) within the group. Each source received 1 piece of gossip and 1 piece of general knowledge from the experimenter during natural class interaction, and the information was allowed to diffuse naturally within the group. Results revealed that gossip was transmitted more frequently than knowledge, and that male sources were more likely to transmit gossip than female sources. The relationship between characteristics of the source, and characteristics of the gossip recipient, also appeared influential with the dominant male source transmitting gossip to exclusively to friends, and the nondominant male source transmitting to individuals of higher peer regard than themselves. Keywords: gossip, open diffusion, source dominance, source sex, transmission biases
Within the developmental and comparative literatures there has been a recent surge of interest in exploring the transmission of behavior across groups of individuals using diffusion techniques (e.g., Flynn & Whiten, 2008; Horner, Whiten, Flynn, & de Waal, 2006; McGuigan & Graham, 2010). Diffusion studies allow the researcher to move away from a dyadic, one-to-one approach, and mimic the creation of cultural traditions at a microlevel. This is achieved by seeding an initial model, preselected by the experimenter, with a behavior of interest, and following the transmission or modification of the behavior across a group of individuals. Diffusion studies take many forms, and have been used regularly in both adult and animal social learning studies (for a review, see Mesoudi & Address correspondence to Nicola McGuigan, Heriot Watt University, School of Life Sciences, Riccarton Campus, Edinburgh, EH14 4AS, UK; [email protected] (e-mail). 605
Downloaded by [University of Connecticut] at 08:01 11 October 2014
606
The Journal of Genetic Psychology
Whiten, 2008; Whiten & Mesoudi, 2008). However, techniques of this nature have rarely been used with children, which is rather surprising as children appear particularly adept at soaking up cultural behavior (e.g., Whiten, McGuigan, Marshall-Pescini, & Hopper, 2009). At present diffusion studies involving children have been almost exclusively confined to a linear diffusion approach, a technique which is structured in much the same way as the children’s game Chinese whispers or telephone, where one seeded expert individual demonstrates a task for individual A, who subsequently demonstrates for individual B, and so on along a chain of individuals (e.g., Flynn, 2008; Flynn & Whiten, 2008; Horner et al., 2006; McGuigan & Graham, 2010). Although an extremely valuable research tool a linear approach loses some of the natural fluidity of transmission as the individuals in the chain, and the order in which they demonstrate, is preselected by the experimenter. In order to circumvent this loss of fluidity Whiten and Flynn (2010) recently conducted the first open diffusion study with children. In open diffusion studies the preselected expert is placed into a group of individuals, and the behavior is allowed to diffuse naturally. This approach to cultural transmission is highly naturalistic as it allows who observes the model, and the order in which observation occurs, to be entirely under the control of the individuals within the group. In their groundbreaking study Whiten and Flynn (2010) trained two expert child models to adopt either a poke or lift technique to obtain a reward from inside an artificial fruit. The experts were then introduced into two different groups of preschool children and the behaviors allowed to spread naturally. The results demonstrated that transmission of the tool actions used by the expert was extensive across both groups, and significantly more children in the experimental groups retrieved the reward than children in a no-model control group, highlighting both the power of cultural learning, and the effectiveness of the open diffusion technique in tracking this transmission. As well as there being a specific lack of open diffusion studies involving children, there is also a general lack of diffusion studies in any domain other than tool use. One as yet unexplored area that is ripe for experimentation is the transmission of social information. The way in which social information is transmitted within groups is of great interest as it may provide insight into the way in which our ancestral social environments shaped the cognition of modern day humans. This social brain (Dunbar, 1998, 2003), or Machiavellian intelligence hypothesis (Byrne & Whiten, 1998; Whiten, 1999; Whiten & Byrne, 1997), suggests that primate intelligence has evolved to tackle complex social problems, with the human-specific social gossip theory (Dunbar, 1993, 1996) suggesting that language, and subsequently gossip, evolved to enable humans to keep track of complex relationships within their social group. In support of these theories, Mesoudi, Whiten, and Dunbar (2006) found that adults of both sexes were more likely to transmit social information than nonsocial information along a chain of
Downloaded by [University of Connecticut] at 08:01 11 October 2014
McGuigan & Cubillo
607
individuals. However, we presently know very little about the emergence of this social bias, a gap, which we aimed to address. In order to explore the transmission of social information Mesoudi et al. (2006) presented participants with a short gossip statement that centered around an illicit affair between a student and a married college professor. Clearly gossip of this nature could not be used with young children, due to the content and the complexity. However, we aimed to capitalize on the tendency shown by children to discuss high authority individuals, in particular teachers (Fine & Rosonow, 1978), by providing pieces of positive gossip about unnamed teachers in their school. The gossip statements were designed to be less intense than that used previously with adults, but still concerned a third party (i.e., an unnamed teacher) engaging in a behavior that was outside the sphere of normal everyday activity. Discussions of high-ranking individuals should be of particular interest as they may provide opportunities for individuals to learn from, and mimic, more successful individuals thereby potentially increasing their own inclusive fitness (Barkow, 1992; De Backer, Neliseen, Vyncke, Braeckman, & McAndrew, 2007; Wert & Salovey, 2004). As well as providing the first empirical exploration of the transmission of social versus nonsocial information in children, the present study also aimed to explore the influence of source characteristics. In the Whiten and Flynn (2010) study the two seeded individuals were both high ranking girls, which begs the question of whether a different pattern of transmission would have been witnessed had the seeded expert been low ranking or boys. There is good reason to suspect that characteristics of the source may be crucial to the way in which information is transmitted, due to the evolution of psychological mechanisms which bias the direction of transmission (Richerson & Boyd, 2005). One such bias has been termed a model-based bias (Richerson & Boyd, 2005), or a context bias (e.g., Henrich & McElreath, 2007), and allows observers to make a choice of what behavior to adopt based on characteristics of the model, with models who are prestigious, competent, successful, or similar to the observer being most influential (Henrich & McElreath, 2007; Richerson & Boyd, 2005). The evidence that presently exists in the developmental literature suggests that dominance plays an important role in the transmission of gossip, but the way it does so is unclear. One set of studies found that high status children played a crucial role in transmitting gossip, particularly with respect to the target and the content of the gossip, (Eder & Encke, 1991; Kuttler, Parker, & La Greca, 2002), whereas Levin and Arluke (1985a) found that it was the isolated and unpopular members of a group who were most likely to gossip in order to enhance their social status. These two apparently contradictory viewpoints may be explained by examining the function of gossip in each case, on the one hand gossip may be used to acquire new friends (Ben-Ze’ev, 1994), on the other hand gossip may be used to facilitate social bonding within an existing group (Dunbar, 2004). We may therefore expect low-ranking individuals to spread
Downloaded by [University of Connecticut] at 08:01 11 October 2014
608
The Journal of Genetic Psychology
gossip to nonfriends, whereas high-ranking individuals may be more likely to gossip with existing friends. With respect to source sex the present data is again somewhat varied. Some studies have shown that women are more frequent gossipers than men (e.g., Kless, 1992; Nevo, Nevo, & Derech-Zehavi, 1993), whereas other studies have found similarities between the sexes (e.g., Levin & Arluke 1985b). What does however appear consistent across studies is (a) that both sexes have a preference for information regarding same sex individuals (McAndrew, Bell, & Garcia, 2007), and (b) men and women tend to vary in the content of their gossip with men being more likely to gossip about sports, celebrities, politicians, and acquaintances, whereas women are more likely to discuss family members and close friends (e.g., Ben-Ze’ev, 1994; Eder & Encke, 1991). These findings appear consistent with sex differences in reproductive tactics where women attempt to gather information about their social network, whereas men self-promote by alluding to their qualities (e.g., sporting prowess; Dunbar, 1996). Overall, in the present study we aimed to adopt an open diffusion approach in order to compare the transmission of gossip and general knowledge within groups of 10- and 11-year-old children. As well as extending our limited knowledge of open diffusion in children to a social information context, we also aimed to explore more fully the influence of the particular sources selected. This was achieved by selecting four sources on two criteria of interest, first their sex (boy or girl), and second their dominance ranking (high or low) within the group. Each source was surreptitiously given one piece of gossip and one piece of general knowledge by the experimenter and the diffusion of the information was allowed to occur naturally. It was predicted that the gossip would be more likely to be transmitted than knowledge within each group, as previous studies have shown a social bias in information transmission (Mesoudi et al., 2006). With respect to the sex and dominance ranking of the information source the predictions were less clear. On the one hand it could be predicted that dominant individuals would be more likely to spread the gossip than nondominant individuals to facilitate social bonding within the group, and on the other hand it was equally plausible that the nondominant individuals would spread gossip as a mechanism to enlarge their social network. Predictions with regard to the sex of the source were also not clear; however, we predicted that if a sex difference was revealed the female sources would be the most likely to spread gossip. Method Participants Forty-four children (20 boys, 24 girls; M age = 11.00 years, SD = 0.66 years) took part in the experiment. The children were recruited from two classes (group 1, n = 20; group 2, n = 24) within a primary school based in central Scotland. Four
McGuigan & Cubillo
609
children, two from each class, took part as the primed information sources. These individuals were selected on the basis of sex and social dominance and included the most dominant boy, the least dominant boy, the most dominant girl, and the least dominant girl.
Downloaded by [University of Connecticut] at 08:01 11 October 2014
Measures Social Dominance Measure. In order to rank each child according to his or her social dominance within their peer group the teacher from each class was asked to “rank the children in your class according to who typically wins disputes and controls resources in the presence of others” (measure adapted from Hawley, 2002). The ranking was linear with the most dominant individual being allocated a score of 1 and the least dominant individual being allocated a score of 20 (group 1) or 24 (group 2). Peer Regard. As well as obtaining a teacher ranking of the relative dominance of each individual we also obtained a measure of peer regard (measure adapted from Hawley, 2002). Peer regard was determined by asking each participant to list the five peers they considered to be the (a) most fun, (b) most likeable, (c) good to play with, (d) a friend, and (e) a best friend. In a sixth section the children were asked to name peers they occasionally fought with. A measure of peer regard was taken for two reasons: to (a) substantiate the rankings provided by the teachers, as previous studies have shown that social dominance is directly related to friendship and popularity within the peer group (Strayer & Trudel, 1984); and (b) allow us to conduct a detailed exploration of the role of friendship in the transmission of gossip and general knowledge. The peer regard questionnaires were initially scored by allocating points to each participant every time they were listed in sections (i–v) of the questionnaire. Five points were allocated for being listed first in a category (i.e., most likeable), four points were allocated for being the second person listed in a category, and so on through to one point for being named as the fifth person considered the most fun, most likeable, and so on. Individuals who were not listed in a category received zero points. Each child was given a peer regard score based on the total number of points they received across the five categories. The minimum score that could be obtained by an individual was zero, indicating that he or she was never listed by any peer in any category. An individual could achieve the maximum score by being listed in first position in all five categories (25 points), by each of his or her classmates (group 1 = 19 peers; group 2 = 23 peers), giving a maximum peer regard score for an individual in group 1 of 475, and for an individual in group 2 a maximum of 575. Children listed in the “peers occasionally fought with” section were deducted one point for each time they were mentioned (e.g., if five peers mentioned that they fought with individual A, individual A’s total score would be deducted five points).
610
The Journal of Genetic Psychology
Downloaded by [University of Connecticut] at 08:01 11 October 2014
Procedure The first phase of the study involved selecting the four information sources on the basis of their social dominance and peer regard scores. The two class teachers completed the social dominance questionnaire (ranking each pupil in their class from most to least dominant), and the children from each class filled out the peer regard questionnaire. For group 1 the peer regard scores ranged from 12 to 116 (M = 57, SD = 29), and for group 2 the scores ranged from four to 160 (M = 61, SD = 39). A comparison of the peer regard and teacher dominance rankings revealed that the individuals who were ranked as the most/least popular within their peer group were also ranked as the most/least dominant by their class teacher, with the two measures correlating significantly for both group 1 (r = .75, p < .001), and group 2 (r = .71, p < .001). This allowed us to select two information sources for group 1 (most dominant boy: teacher ranking = 1, peer regard score = 116; the least dominant boy: teacher ranking = 20, peer regard score = 12), and two information sources for group 2 (most dominant girl: teacher ranking = 1, peer regard score = 160; the least dominant girl: teacher ranking = 24, peer regard score = 4). Once the information sources had been identified the experimenter used a natural point in the classroom activities to drop both a gossip statement (e.g., “A teacher at your school is auditioning for X-factor”) and a general knowledge statement (e.g., “It is impossible to sneeze with your eyes open”) into general conversation with each source (see Table 1 for complete list of gossip and knowledge statements). The source was alone with the experimenter during the information
TABLE 1. Gossip and Knowledge Statements as a Function of Information Source Source Most dominant boy Least dominant boy Most dominant girl Least dominant girl
Gossip statement
Knowledge statement
A teacher at your school is training for the London Olympics A teacher at your school is buying a Porsche
There are around 61 million people living in the United Kingdom Chinese is the most widely spoken language in the world Lipstick is partially made of fish scales It is impossible to sneeze with your eyes open
A teacher at your school is auditioning for x-factor A teacher at your school is getting a pet monkey for Christmas
McGuigan & Cubillo
611
Downloaded by [University of Connecticut] at 08:01 11 October 2014
transfer, and there was no instruction to the source to pass the information on, with the result that any subsequent diffusion was spontaneously initiated by the source. In order to determine whether the information had diffused the experimenter returned to the school two days after the information had been passed to the source and asked each child to complete a short questionnaire asking whether he or she had heard (yes or no) the four gossip statements and the four general knowledge statements. If the children indicated that they had heard the information they were asked to list who they had heard the statements from. Results Transmission of Information: Gossip Versus Knowledge Of initial interest was whether the gossip statements were transmitted with more frequency within the groups than the knowledge statements. In order to explore this possibility, the number of individuals who reported hearing the gossip statements and the knowledge statements were totaled across the two groups. The gossip statements (heard by 22 individuals) were found to have spread among the groups with significantly greater frequency than the knowledge statements (heard by eight individuals; Fisher’s exact test, p = .016; see Figure 1), suggesting a social bias in transmission. However, there were differences in the extent to which gossip was transmitted within each group. In group 1 the gossip transmitted extensively, with 18 individuals reporting having heard the gossip statements, as compared to only three individuals who indicated that they had heard the knowledge statements (Fisher’s exact test, p = .001). Unfortunately equivalent statistical comparisons could not be made with respect to group 2 as the transmission of both gossip (heard by four individuals), and knowledge (heard by five individuals) was infrequent in group 2. Characteristics of the Information Source Dominance of Source. As well as exploring whether gossip transmitted more readily than knowledge we were also interested in the impact of source dominance on transmission. In order to explore the influence of dominance we first of all totaled the number of individuals who reported hearing the gossip and knowledge statements directly from the two nondominant sources and the two dominant sources, before going on to explore each of the four sources individually. This revealed that the two dominant sources transmitted the gossip to a total of 10 individuals and the two nondominant sources transmitted the gossip to seven individuals, a difference which was not significant. This pattern of results was mirrored in the performance of the individual sources as little difference was evident between the two male sources (dominant boy told eight individuals; nondominant boy told six individuals), or the two female sources (dominant girl told two individuals; nondominant
Downloaded by [University of Connecticut] at 08:01 11 October 2014
612
The Journal of Genetic Psychology
FIGURE 1. The left-hand side of the figure shows the transmission of gossip and general knowledge in group 1, and the right-hand side shows transmission in group 2. A represents the dominant boy, B represents the nondominant boy, C represents the dominant girl, and D represents the nondominant female. The transmission of gossip is represented with an arrow and the transmission of general knowledge is represented by a circle. Friends are shown with a checked background and nonfriends with a spotted background. M = male recipient, F = female recipient.
McGuigan & Cubillo
613
Downloaded by [University of Connecticut] at 08:01 11 October 2014
girl told one individual), suggesting that dominance level had little impact on gossip transmission. Similarly, the dominance level of the source had little influence on the transmission of knowledge, either across the source type combined (nondominant sources transmitted to three individuals; dominant sources transmitted to five individuals), or at the level of the individual source (dominant boy told one individual; nondominant boy told one individual; dominant girl told four individuals; nondominant girl told two individuals). Sex of Source. The second characteristic of the source that may have influenced the spread of the information was the sex of the source. As the dominance level of the source had little impact on transmission we combined the data from the two male sources and the two female sources to explore the influence of source sex. This revealed that the male sources transmitted the gossip directly to 14 individuals, whereas the two female sources spread the gossip to only three individuals, a difference which was found to be significant (Fisher’s exact test, p = .013). Unfortunately due to the small sample size equivalent statistical comparisons could not be made with respect to the transmission of knowledge, however the trend was toward greater transmission by the female sources (male sources told two individuals; female sources told six individuals). Characteristics of Information Recipient Friend or Nonfriend. As each participant provided a list of friends we could use this information to explore whether information was more readily transmitted to friends or nonfriends. Across the two groups the gossip statements spread to a total of 22 individuals, 13 friends and nine nonfriends, a difference which was not significant. Although the small sample size does not allow for detailed statistical analysis of the individual sources there appeared to be trends in the data, with the dominant male transmitting gossip exclusively to friends (eight friends vs. zero nonfriends), and the nondominant boy transmitting to an equal number of friends and nonfriends (three individuals in each case). The female sources transmitted very little gossip overall but that which did occur was as likely to be a friend (two individuals), or a nonfriend (one individual). In contrast the pattern with respect to the knowledge statements was somewhat different, with the participants indicating that they heard the knowledge statements exclusively from friends rather than nonfriends (eight friends vs. zero nonfriends). Dominance of Recipient. An additional factor that may play a role in the transfer of gossip is the recipient’s peer standing within the group, and their position in relation to the dominance level of the source. In order to explore this relationship we averaged the peer regard scores of the 10 individuals who received the gossip from the two dominant sources, and the seven individuals who received the gossip from the two nondominant sources. This revealed that the peer regard scores of the individuals who received the gossip from the nondominant sources were
614
The Journal of Genetic Psychology
Downloaded by [University of Connecticut] at 08:01 11 October 2014
significantly higher (M peer regard = 86, SD = 33), than those of the individuals who received the gossip from the dominant sources (M peer regard = 68, SD = 23), F(1, 15) = 107, p < .001, μ2 = .097. This was particularly true of the least dominant boy who transmitted the gossip to individuals of a significantly higher social standing (M peer regard score = 89, SD = 31), than the dominant boy (M peer regard score = 68, SD = 24), the dominant girl (M peer regard score = 65, SD = 17), and the nondominant girl (M peer regard score = 18, SD = 0; p < .001 in each case). Sex of Recipient. The final question of interest was whether individuals would be more likely to transmit information to individuals of the same sex as themselves. Across the two groups there was a nonsignificant trend toward the transmission of gossip to same-sex individuals (15 occasions), than opposite-sex individuals (seven occasions). With respect to the individual sources the limited sample size does not allow for statistical analysis. However, the trends in the data appeared to show little variation in the sex of the gossip recipient with the exception of the dominant boy who transmitted to same-sex individuals (six individuals) more frequently than opposite-sex individuals (two individuals). In contrast, the sex of the recipient had very little impact on the transfer of knowledge as the information was transmitted to three same-sex individuals, and five opposite-sex individuals. Discussion The present study used an open diffusion approach to explore the transmission of gossip and general knowledge within two groups of 10- and 11-year-old children. The groups were seeded by two information sources preselected on the basis of their sex, and dominance ranking within their peer group. The results indicated that the type of information was crucial to transmission, with gossip spreading among the children more readily than general knowledge, thus demonstrating a social bias in information transmission earlier than previously witnessed. However, extensive transmission of the gossip occurred in one group only, suggesting that particular characteristics of the sources may have impacted on transmission. One characteristic of the source that seemed particularly influential was the sex of the source, with the male sources transmitting more gossip than the female sources. This finding is somewhat counter to previous literature which tends to report more frequent gossiping by females (e.g., Kless, 1992; Nevo et al., 1993). Of interest is why the male sources transmitted gossip so readily? Conversely why were the female sources more restrained in this respect? The first possibility lies in the content of the gossip itself. The type of information which may be of most interest to males is information that concerns the status of another male, including information that is relevant to resources and physical prowess (McAndrew & Milenkovic, 2002). Arguably the two gossip statements provided to the male sources contained information which was more likely to be about a same sex
Downloaded by [University of Connecticut] at 08:01 11 October 2014
McGuigan & Cubillo
615
other (male teacher), as well as alluding to the gossipee’s resources (Porsche), or physical ability (Olympics). This attribution may have been less likely for the statements provided to the female sources, which could have been perceived as more gender neutral. Perhaps the neutrality of the statements, and the subsequent lack of gender specificity, was less appealing to the female sources. Alternatively, the female sources might have failed to transmit the gossip as the gossip statements did not refer to relationships between individuals. Many studies have shown that a key characteristic of female gossip, particularly female–female gossip, is that it concerns relationships, especially those of family and friends (e.g., Levin & Arluke, 1985b). The gossip provided here lacked such a relational component, potentially downplaying its interest to the female sources. A second explanation for the sex difference witnessed may lie in the different structure of male and female social groups during this age period. Boys of this age tend to have loose, and extensive social groups, whereas girls form tighter social groups, often pair bonds during this period (Lagerspetz, Bj¨orkqvist, & Peltonen, 1988). The loose nature of the male social group may lend itself to a more widespread transmission of gossip, whereas tighter social groups may lead to reduced transmission to a select few. The differences may also be explained by the different ways that aggression is used by each sex, with boys tending to employ more direct overt aggression (e.g., hitting), and girls using more indirect (e.g., negative gossip, back-biting) relational aggression (Bj¨orkqvist, Lagerspetz, & Kaukianen, 1992; Crick, 1996; Crick & Grotpeter, 1995). The greater importance of negative gossip for girls has led to the suggestion that gossip is key to maintaining status in female groups (Eckert, 1990; Leaper & Holliday, 1995), with the result that young females may be more attuned to the truthfulness of gossip than young males (Kuttler et al., 2002). This might have resulted in the female sources being more skeptical about the truthfulness of the gossip, and subsequently more reluctant to transmit the information to other individuals. As well as exploring the role of source characteristics we were also interested in who the information was transmitted to, as well as the relationship between the gossiper and the gossip recipient. The first question of interest was whether the information was as likely to be transmitted to a friend or a nonfriend. Previous research has been somewhat contradictory in this respect, with some studies suggesting that gossip is more likely to occur between friends (Leaper & Holliday, 1995), and other studies suggesting that children are as likely to gossip with friends as they are acquaintances (Mettetal, 1983). The present study suggests that there may be merit in both claims depending on the function of gossip in each case. Looking across the two groups we see a nonsignificant trend toward transmission of gossip to friends. However, when looking at the sources individually the dominant boy transmitted exclusively to friends whereas the nondominant boy, whereas both of the female sources were as likely to transmit to friends as they were nonfriends. This finding supports the suggestion that the transmission of gossip may have been serving different functions in each case. One key feature of gossip is
Downloaded by [University of Connecticut] at 08:01 11 October 2014
616
The Journal of Genetic Psychology
that it can serve to strengthen the social bonds with a group (Ben-Ze’ev, 1994). The dominant boy is likely to have an extensive social network, and by passing the gossip to members of this network the group bonds may have been strengthened. In contrast, the position may be somewhat different for the nondominant boy, who is likely to have a smaller social network, and may have been attempting to use the gossip as a tool to acquire friendships (McAndrew et al., 2007). In addition to using gossip as a tool to acquire new friends it appeared that the nondominant boy might have been attempting to use gossip strategically in order to enhance his social status within the group. Barkow (1992) suggested that from the preschool years onward individuals generally attempt to improve our social standing with our group. Social dominance is extremely desirable as dominant individuals are viewed as more attractive social partners (LaFreniere & Charlesworth, 1983), are looked at (Hawley & Little, 1999; Vaughn & Waters, 1981), and imitated more by peers (Abramovitch & Grusec, 1978), and have greater success in controlling resources (Hawley, 2002). One mechanism an individual could use in an attempt to enhance social status is to impart novel social information to a more dominant, sought after, member of the social group (McAndrew et al., 2007). This claim is supported by the tendency of the low-ranking boy to spread gossip to individuals with a higher peer regard score than did the dominant boy. Overall, the present results provide the first empirical evidence of a social bias the transmission of information in childhood. However, the results appear to suggest that the transmission of social information is multifaceted and highly dependent on many different factors including the characteristics of the gossiper, characteristics of the gossip recipient, and the relationship between the two in terms of friendship and relative social standing. This complexity could be explained from two, potentially intimately related, perspectives. One possible explanation may lie at the level of the child’s social world, in particular from the perspective of social learning. More specifically, within the present study social learning may be evident in relation to the content of the information as preferential attention to the gossip statements may have provided the children with potentially useful information about successful, salient individuals (i.e., teachers) in their social world. Similarly, a social learning explanation could also account for the pattern of transmission witnessed within the peer group itself. Abramovitch and Grusec (1978) showed that children in the present age range who are rated as most dominant within their group are looked at and imitated more frequently than nondominant individuals. A potential explanation for this differential treatment is that dominant individuals are seen as prized sources of knowledge, leading other individuals in the group to monitor their behavior closely so that they can learn appropriate behaviors from them. If we combine this social learning tendency with the finding that boys participate in larger social groups than girls, then it would appear that boys have greater opportunity to take an interest in and attend to the behavior of other individuals. This increased level of opportunity may explain the greater levels of transmission by the dominant male source in the present study.
McGuigan & Cubillo
617
Downloaded by [University of Connecticut] at 08:01 11 October 2014
Intriguingly the patterns of performance witnessed also fit well with the theory that the way in which information is transmitted within groups is underpinned by evolved psychological mechanisms (e.g., Richerson & Boyd, 2005) which bias the transmission of information both at the level of the content of the information (content-based bias) and at the level of the information provider (model-based bias). These intriguing findings, although preliminary, pave the way for more detailed explorations of the relationship between source characteristics and the content of the information to be transmitted. AUTHOR NOTES Nicola McGuigan is a lecturer in the department of psychology, School of Life Sciences at Heriot Watt University in Scotland. Her research interests lie in social learning and more recently in prosocial behavior. Marcus Cubillo is a researcher in the department of psychology, School of Life Sciences at Heriot Watt University. His research interests include information transmission within groups. REFERENCES Abramovitch, R., & Grusec, J. E. (1978). Peer imitation in a natural setting. Child Development, 49, 60–65. Barkow, J. H. (1992). Beneath new culture is old psychology: Gossip and social stratification. In J. H. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 627–637). New York, NY: Oxford University Press. Ben-Ze’ev, A. (1994). The vindication of gossip. In R. F. Goodman & A. Ben-Ze’ev (Eds.), Good gossip (pp. 11–24). Lawrence, KS: University Press of Kansas. Bj¨orkqvist, K., Lagerspetz, K. M. J., & Kaukianen, A. (1992). Do girls manipulate and boys fight? Developmental trends in regard to direct and indirect aggression. Aggressive Behavior, 18, 117–127. Byrne, R. W., & Whiten, A. (1988). Machiavellian intelligence-social expertise and the evolution of intellect in monkeys, apes and humans. Oxford, England: Claredon. Crick, N. R. (1996). The role of overt aggression, relational aggression, and prosocial behavior in the prediction of children’s future social adjustment. Child Development, 67, 2317–2327. Crick, N. R., & Grotpeter, J. K. (1995). Relational aggression, gender, and socialpsychological adjustment. Child Development, 66, 710–722. De Backer, C., Neliseen, M., Vyncke, P., Braeckman, J., & McAndrew, F. (2007). Celebrities: From teachers to friends. Human Nature, 18, 334–354. Dunbar, R. I. M. (1993). Coevolution of neocortical size, group size and language. Behavioral and Brain Sciences, 16, 681–735. Dunbar, R. I. M. (1996). Grooming, gossip and the evolution of language. London, England: Faber and Faber. Dunbar, R. I. M. (1998). The social brain hypothesis. Evolutionary Anthropology, 6, 178–190. Dunbar, R. I. M. (2003). The social brain: Mind, language, and society in evolutionary perspective. Annual Review of Anthropology, 32, 163–181.
Downloaded by [University of Connecticut] at 08:01 11 October 2014
618
The Journal of Genetic Psychology
Dunbar, R. I. M. (2004). Gossip in evolutionary perspective. Review of General Psychology, 8, 100–110. Eckert, P. (1990). Cooperative competition in adolescent “girl talk.” Discourse Processes, 13, 91–122. Eder, D., & Enke, J. L. (1991). The structure of gossip: Opportunities and constraints on collective expression among adolescents. American Sociological Review, 56, 494–508. Fine, G. A., & Rosnow, R. L. (1978). Gossip, gossipers, gossiping. Personality and Social Psychology Bulletin, 4, 161–168. Flynn, E. (2008). Investigating children as cultural magnets: do young children transmit redundant information along diffusion chains? Philosophical Transmissions of The Royal Society, 363, 3541–3551. Flynn, E., & Whiten, A. (2008) Cultural transmission of tool use in young children: A diffusion chain study. Social Development, 17, 699–718. Hawley, P. H. (2002). Social dominance and prosocial and coercive strategies of resource control in preschoolers. International Journal of Behavioral Development, 26, 167–176. Hawley, P. H., & Little, T. D. (1999). Winning some and losing some: A social relations approach to social dominance in toddlers. Merrill-Palmer Quarterly, 45, 185–214. Henrich, J., & McElreath, R. (2007). Dual-inheritance theory: The evolution of human cultural capacities and cultural evolution. In R. Dunbar & L. Barrett (Eds.), Oxford handbook of evolutionary psychology (pp. 555–570). Oxford, England: Oxford University Press. Horner, V., Whiten, A., Flynn, E., & de Waal, F. B. M. (2006). Faithful replication of foraging techniques along cultural transmission chains by chimpanzees and children. National Academic Sciences, 103, 13878–13883. Kless, S. J. (1992). The attainment of peer status: Gender and power relationships in the elementary school. Sociological Studies of Child Development, 5, 115–148. Kuttler, A. F., Parker, J. G., & La Greca, A. M. (2002). Developmental and gender differences in preadolescents’ judgments of the veracity of gossip. Merrill-Palmer Quarterly, 48, 105–132. LaFreniere, P. J., & Charlesworth, W. R. (1987). Effects of friendship and dominance status on preschooler’s resource utilization in a cooperative/competitive paradigm. International Journal of Behavioral Development, 10, 345–358. Lagerspetz, K. M. J., Bj¨orkqvist, K., & Peltonen, T. (1988). Is indirect aggression typical of females? Gender differences in aggressiveness in 11- to 12-year-old children. Aggressive Behavior, 14, 403–414. Leaper, C., & Holliday, H. (1995). Gossip in a same-gender and cross gender friends’ conversations. Personal Relationships, 2, 237–246. Levin, J., & Arluke, A. (1985a). Gossip: The inside scoop. New York, NY: Plenum. Levin, J., & Arluke, A. (1985b). An exploratory analysis of sex differences in gossip. Sex Roles, 12, 281–286. McAndrew, F. T., Bell, E. K., & Garcia, C. M. (2007). Who do we tell and whom do we tell on? Gossip as a strategy for status enhancement. Journal of Applied Social Psychology, 37, 1562–1577. McAndrew, F. T., & Milenkovic, M. A. (2002). Of tabloids and family secrets: The evolutionary psychology of gossip. Journal of Applied Social Psychology, 32, 1064–1082. McGuigan, N., & Graham, M. (2010). Cultural transmission of irrelevant tool actions in diffusion chains of 3- and 5-year-old children. European Journal of Developmental Psychology, 7, 561–577. Mesoudi, A., & Whiten, A. (2008). Establishing an experimental science of culture: Animal social diffusion experiments. Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences, 363, 3489–3502.
Downloaded by [University of Connecticut] at 08:01 11 October 2014
McGuigan & Cubillo
619
Mesoudi, A., Whiten, A., & Dunbar, R. I. M. (2006). A bias for social information in human cultural transmission. British Journal of Psychology, 97, 405–423. Mettetal, G. (1983). Fantasy, gossip, and self-disclosure: Children’s conversations with friends. Communication Yearbook, 7, 717–735. Nevo, O., Nevo, B., & Derech-Zehavi, A. (1993). The development of the tendency to gossip. questionnaire: Construct and concurrent validity for a sample of Israeli college students. Educational and Psychological Measurement, 53, 973–981. Richerson, P. J., & Boyd, R. (2005). Not by genes alone. Chicago, IL: University of Chicago Press. Strayer, F. F., & Trudel, M. (1984). Developmental changes in the nature and function of social dominance among young children. Ethology and Sociobiology, 5, 279–295. Vaughn, B. E., & Waters, E. (1981). Attention structure, sociometric status, and dominance: Interrelations, behavioral correlates, and relationships to social competence. Developmental Psychology, 17, 275–288. Wert, S. R., & Salovey, P. (2004). A social comparison account of gossip. Review of General Psychology, 8, 122–137. Whiten, A. (1999). The evolution of deep social mind in humans. In M. C. Corballis & S. E. G. Lea (Eds.), The descent of mind: Psychological perspectives on hominid evolution (pp. 173–193). Oxford, England: Oxford University Press. Whiten, A., & Byrne, R. W. (Eds.). (1997). Machiavellian intelligence II: Extensions and evaluations. New York, NY: Cambridge University Press. Whiten, A., & Flynn, E. G. (2010). The transmission and evolution of experimental microcultures in groups of young children. Developmental Psychology, 46, 1694–1709. Whiten, A., McGuigan, N., Marshall-Pescini, S., & Hopper, L. (2009). Emulation, imitation, over-imitation and the scope of culture for child and chimpanzee. Philosophical Transactions of the Royal Society B, 364, 2117–2428. Whiten, A., & Mesoudi, A. (2008). An experimental science of culture: Animal social diffusion experiments. Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences, 363, 3477–3488.
Original manuscript received July 16, 2012 Final version accepted November 13, 2012
The Journal of Genetic Psychology: Research and Theory on Human Development Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/vgnt20
Information Transmission in Young Children: When Social Information Is More Important Than Nonsocial Information a
Nicola McGuigan & Marcus Cubillo
a
a
Heriot-Watt University , Scotland Accepted author version posted online: 24 Apr 2013.Published online: 02 Oct 2013.
To cite this article: Nicola McGuigan & Marcus Cubillo (2013) Information Transmission in Young Children: When Social Information Is More Important Than Nonsocial Information, The Journal of Genetic Psychology: Research and Theory on Human Development, 174:6, 605-619, DOI: 10.1080/00221325.2012.749833 To link to this article: http://dx.doi.org/10.1080/00221325.2012.749833
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages,
and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.
Downloaded by [University of Connecticut] at 08:01 11 October 2014
This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions
Downloaded by [University of Connecticut] at 08:01 11 October 2014
The Journal of Genetic Psychology, 2013, 174(6), 605–619 C Taylor & Francis Group, LLC Copyright doi: 10.1080/00221325.2012.749833
Information Transmission in Young Children: When Social Information Is More Important Than Nonsocial Information NICOLA McGUIGAN MARCUS CUBILLO Heriot-Watt University, Scotland
ABSTRACT. The authors’ aim was to use a highly novel open diffusion paradigm to investigate the transmission of social information (i.e., gossip) and general knowledge within 2 groups of 10- and 11-year-old children. Four children, 2 from each group, acted as a primed information source, selected on the basis of sex and dominance ranking (high or low) within the group. Each source received 1 piece of gossip and 1 piece of general knowledge from the experimenter during natural class interaction, and the information was allowed to diffuse naturally within the group. Results revealed that gossip was transmitted more frequently than knowledge, and that male sources were more likely to transmit gossip than female sources. The relationship between characteristics of the source, and characteristics of the gossip recipient, also appeared influential with the dominant male source transmitting gossip to exclusively to friends, and the nondominant male source transmitting to individuals of higher peer regard than themselves. Keywords: gossip, open diffusion, source dominance, source sex, transmission biases
Within the developmental and comparative literatures there has been a recent surge of interest in exploring the transmission of behavior across groups of individuals using diffusion techniques (e.g., Flynn & Whiten, 2008; Horner, Whiten, Flynn, & de Waal, 2006; McGuigan & Graham, 2010). Diffusion studies allow the researcher to move away from a dyadic, one-to-one approach, and mimic the creation of cultural traditions at a microlevel. This is achieved by seeding an initial model, preselected by the experimenter, with a behavior of interest, and following the transmission or modification of the behavior across a group of individuals. Diffusion studies take many forms, and have been used regularly in both adult and animal social learning studies (for a review, see Mesoudi & Address correspondence to Nicola McGuigan, Heriot Watt University, School of Life Sciences, Riccarton Campus, Edinburgh, EH14 4AS, UK; [email protected] (e-mail). 605
Downloaded by [University of Connecticut] at 08:01 11 October 2014
606
The Journal of Genetic Psychology
Whiten, 2008; Whiten & Mesoudi, 2008). However, techniques of this nature have rarely been used with children, which is rather surprising as children appear particularly adept at soaking up cultural behavior (e.g., Whiten, McGuigan, Marshall-Pescini, & Hopper, 2009). At present diffusion studies involving children have been almost exclusively confined to a linear diffusion approach, a technique which is structured in much the same way as the children’s game Chinese whispers or telephone, where one seeded expert individual demonstrates a task for individual A, who subsequently demonstrates for individual B, and so on along a chain of individuals (e.g., Flynn, 2008; Flynn & Whiten, 2008; Horner et al., 2006; McGuigan & Graham, 2010). Although an extremely valuable research tool a linear approach loses some of the natural fluidity of transmission as the individuals in the chain, and the order in which they demonstrate, is preselected by the experimenter. In order to circumvent this loss of fluidity Whiten and Flynn (2010) recently conducted the first open diffusion study with children. In open diffusion studies the preselected expert is placed into a group of individuals, and the behavior is allowed to diffuse naturally. This approach to cultural transmission is highly naturalistic as it allows who observes the model, and the order in which observation occurs, to be entirely under the control of the individuals within the group. In their groundbreaking study Whiten and Flynn (2010) trained two expert child models to adopt either a poke or lift technique to obtain a reward from inside an artificial fruit. The experts were then introduced into two different groups of preschool children and the behaviors allowed to spread naturally. The results demonstrated that transmission of the tool actions used by the expert was extensive across both groups, and significantly more children in the experimental groups retrieved the reward than children in a no-model control group, highlighting both the power of cultural learning, and the effectiveness of the open diffusion technique in tracking this transmission. As well as there being a specific lack of open diffusion studies involving children, there is also a general lack of diffusion studies in any domain other than tool use. One as yet unexplored area that is ripe for experimentation is the transmission of social information. The way in which social information is transmitted within groups is of great interest as it may provide insight into the way in which our ancestral social environments shaped the cognition of modern day humans. This social brain (Dunbar, 1998, 2003), or Machiavellian intelligence hypothesis (Byrne & Whiten, 1998; Whiten, 1999; Whiten & Byrne, 1997), suggests that primate intelligence has evolved to tackle complex social problems, with the human-specific social gossip theory (Dunbar, 1993, 1996) suggesting that language, and subsequently gossip, evolved to enable humans to keep track of complex relationships within their social group. In support of these theories, Mesoudi, Whiten, and Dunbar (2006) found that adults of both sexes were more likely to transmit social information than nonsocial information along a chain of
Downloaded by [University of Connecticut] at 08:01 11 October 2014
McGuigan & Cubillo
607
individuals. However, we presently know very little about the emergence of this social bias, a gap, which we aimed to address. In order to explore the transmission of social information Mesoudi et al. (2006) presented participants with a short gossip statement that centered around an illicit affair between a student and a married college professor. Clearly gossip of this nature could not be used with young children, due to the content and the complexity. However, we aimed to capitalize on the tendency shown by children to discuss high authority individuals, in particular teachers (Fine & Rosonow, 1978), by providing pieces of positive gossip about unnamed teachers in their school. The gossip statements were designed to be less intense than that used previously with adults, but still concerned a third party (i.e., an unnamed teacher) engaging in a behavior that was outside the sphere of normal everyday activity. Discussions of high-ranking individuals should be of particular interest as they may provide opportunities for individuals to learn from, and mimic, more successful individuals thereby potentially increasing their own inclusive fitness (Barkow, 1992; De Backer, Neliseen, Vyncke, Braeckman, & McAndrew, 2007; Wert & Salovey, 2004). As well as providing the first empirical exploration of the transmission of social versus nonsocial information in children, the present study also aimed to explore the influence of source characteristics. In the Whiten and Flynn (2010) study the two seeded individuals were both high ranking girls, which begs the question of whether a different pattern of transmission would have been witnessed had the seeded expert been low ranking or boys. There is good reason to suspect that characteristics of the source may be crucial to the way in which information is transmitted, due to the evolution of psychological mechanisms which bias the direction of transmission (Richerson & Boyd, 2005). One such bias has been termed a model-based bias (Richerson & Boyd, 2005), or a context bias (e.g., Henrich & McElreath, 2007), and allows observers to make a choice of what behavior to adopt based on characteristics of the model, with models who are prestigious, competent, successful, or similar to the observer being most influential (Henrich & McElreath, 2007; Richerson & Boyd, 2005). The evidence that presently exists in the developmental literature suggests that dominance plays an important role in the transmission of gossip, but the way it does so is unclear. One set of studies found that high status children played a crucial role in transmitting gossip, particularly with respect to the target and the content of the gossip, (Eder & Encke, 1991; Kuttler, Parker, & La Greca, 2002), whereas Levin and Arluke (1985a) found that it was the isolated and unpopular members of a group who were most likely to gossip in order to enhance their social status. These two apparently contradictory viewpoints may be explained by examining the function of gossip in each case, on the one hand gossip may be used to acquire new friends (Ben-Ze’ev, 1994), on the other hand gossip may be used to facilitate social bonding within an existing group (Dunbar, 2004). We may therefore expect low-ranking individuals to spread
Downloaded by [University of Connecticut] at 08:01 11 October 2014
608
The Journal of Genetic Psychology
gossip to nonfriends, whereas high-ranking individuals may be more likely to gossip with existing friends. With respect to source sex the present data is again somewhat varied. Some studies have shown that women are more frequent gossipers than men (e.g., Kless, 1992; Nevo, Nevo, & Derech-Zehavi, 1993), whereas other studies have found similarities between the sexes (e.g., Levin & Arluke 1985b). What does however appear consistent across studies is (a) that both sexes have a preference for information regarding same sex individuals (McAndrew, Bell, & Garcia, 2007), and (b) men and women tend to vary in the content of their gossip with men being more likely to gossip about sports, celebrities, politicians, and acquaintances, whereas women are more likely to discuss family members and close friends (e.g., Ben-Ze’ev, 1994; Eder & Encke, 1991). These findings appear consistent with sex differences in reproductive tactics where women attempt to gather information about their social network, whereas men self-promote by alluding to their qualities (e.g., sporting prowess; Dunbar, 1996). Overall, in the present study we aimed to adopt an open diffusion approach in order to compare the transmission of gossip and general knowledge within groups of 10- and 11-year-old children. As well as extending our limited knowledge of open diffusion in children to a social information context, we also aimed to explore more fully the influence of the particular sources selected. This was achieved by selecting four sources on two criteria of interest, first their sex (boy or girl), and second their dominance ranking (high or low) within the group. Each source was surreptitiously given one piece of gossip and one piece of general knowledge by the experimenter and the diffusion of the information was allowed to occur naturally. It was predicted that the gossip would be more likely to be transmitted than knowledge within each group, as previous studies have shown a social bias in information transmission (Mesoudi et al., 2006). With respect to the sex and dominance ranking of the information source the predictions were less clear. On the one hand it could be predicted that dominant individuals would be more likely to spread the gossip than nondominant individuals to facilitate social bonding within the group, and on the other hand it was equally plausible that the nondominant individuals would spread gossip as a mechanism to enlarge their social network. Predictions with regard to the sex of the source were also not clear; however, we predicted that if a sex difference was revealed the female sources would be the most likely to spread gossip. Method Participants Forty-four children (20 boys, 24 girls; M age = 11.00 years, SD = 0.66 years) took part in the experiment. The children were recruited from two classes (group 1, n = 20; group 2, n = 24) within a primary school based in central Scotland. Four
McGuigan & Cubillo
609
children, two from each class, took part as the primed information sources. These individuals were selected on the basis of sex and social dominance and included the most dominant boy, the least dominant boy, the most dominant girl, and the least dominant girl.
Downloaded by [University of Connecticut] at 08:01 11 October 2014
Measures Social Dominance Measure. In order to rank each child according to his or her social dominance within their peer group the teacher from each class was asked to “rank the children in your class according to who typically wins disputes and controls resources in the presence of others” (measure adapted from Hawley, 2002). The ranking was linear with the most dominant individual being allocated a score of 1 and the least dominant individual being allocated a score of 20 (group 1) or 24 (group 2). Peer Regard. As well as obtaining a teacher ranking of the relative dominance of each individual we also obtained a measure of peer regard (measure adapted from Hawley, 2002). Peer regard was determined by asking each participant to list the five peers they considered to be the (a) most fun, (b) most likeable, (c) good to play with, (d) a friend, and (e) a best friend. In a sixth section the children were asked to name peers they occasionally fought with. A measure of peer regard was taken for two reasons: to (a) substantiate the rankings provided by the teachers, as previous studies have shown that social dominance is directly related to friendship and popularity within the peer group (Strayer & Trudel, 1984); and (b) allow us to conduct a detailed exploration of the role of friendship in the transmission of gossip and general knowledge. The peer regard questionnaires were initially scored by allocating points to each participant every time they were listed in sections (i–v) of the questionnaire. Five points were allocated for being listed first in a category (i.e., most likeable), four points were allocated for being the second person listed in a category, and so on through to one point for being named as the fifth person considered the most fun, most likeable, and so on. Individuals who were not listed in a category received zero points. Each child was given a peer regard score based on the total number of points they received across the five categories. The minimum score that could be obtained by an individual was zero, indicating that he or she was never listed by any peer in any category. An individual could achieve the maximum score by being listed in first position in all five categories (25 points), by each of his or her classmates (group 1 = 19 peers; group 2 = 23 peers), giving a maximum peer regard score for an individual in group 1 of 475, and for an individual in group 2 a maximum of 575. Children listed in the “peers occasionally fought with” section were deducted one point for each time they were mentioned (e.g., if five peers mentioned that they fought with individual A, individual A’s total score would be deducted five points).
610
The Journal of Genetic Psychology
Downloaded by [University of Connecticut] at 08:01 11 October 2014
Procedure The first phase of the study involved selecting the four information sources on the basis of their social dominance and peer regard scores. The two class teachers completed the social dominance questionnaire (ranking each pupil in their class from most to least dominant), and the children from each class filled out the peer regard questionnaire. For group 1 the peer regard scores ranged from 12 to 116 (M = 57, SD = 29), and for group 2 the scores ranged from four to 160 (M = 61, SD = 39). A comparison of the peer regard and teacher dominance rankings revealed that the individuals who were ranked as the most/least popular within their peer group were also ranked as the most/least dominant by their class teacher, with the two measures correlating significantly for both group 1 (r = .75, p < .001), and group 2 (r = .71, p < .001). This allowed us to select two information sources for group 1 (most dominant boy: teacher ranking = 1, peer regard score = 116; the least dominant boy: teacher ranking = 20, peer regard score = 12), and two information sources for group 2 (most dominant girl: teacher ranking = 1, peer regard score = 160; the least dominant girl: teacher ranking = 24, peer regard score = 4). Once the information sources had been identified the experimenter used a natural point in the classroom activities to drop both a gossip statement (e.g., “A teacher at your school is auditioning for X-factor”) and a general knowledge statement (e.g., “It is impossible to sneeze with your eyes open”) into general conversation with each source (see Table 1 for complete list of gossip and knowledge statements). The source was alone with the experimenter during the information
TABLE 1. Gossip and Knowledge Statements as a Function of Information Source Source Most dominant boy Least dominant boy Most dominant girl Least dominant girl
Gossip statement
Knowledge statement
A teacher at your school is training for the London Olympics A teacher at your school is buying a Porsche
There are around 61 million people living in the United Kingdom Chinese is the most widely spoken language in the world Lipstick is partially made of fish scales It is impossible to sneeze with your eyes open
A teacher at your school is auditioning for x-factor A teacher at your school is getting a pet monkey for Christmas
McGuigan & Cubillo
611
Downloaded by [University of Connecticut] at 08:01 11 October 2014
transfer, and there was no instruction to the source to pass the information on, with the result that any subsequent diffusion was spontaneously initiated by the source. In order to determine whether the information had diffused the experimenter returned to the school two days after the information had been passed to the source and asked each child to complete a short questionnaire asking whether he or she had heard (yes or no) the four gossip statements and the four general knowledge statements. If the children indicated that they had heard the information they were asked to list who they had heard the statements from. Results Transmission of Information: Gossip Versus Knowledge Of initial interest was whether the gossip statements were transmitted with more frequency within the groups than the knowledge statements. In order to explore this possibility, the number of individuals who reported hearing the gossip statements and the knowledge statements were totaled across the two groups. The gossip statements (heard by 22 individuals) were found to have spread among the groups with significantly greater frequency than the knowledge statements (heard by eight individuals; Fisher’s exact test, p = .016; see Figure 1), suggesting a social bias in transmission. However, there were differences in the extent to which gossip was transmitted within each group. In group 1 the gossip transmitted extensively, with 18 individuals reporting having heard the gossip statements, as compared to only three individuals who indicated that they had heard the knowledge statements (Fisher’s exact test, p = .001). Unfortunately equivalent statistical comparisons could not be made with respect to group 2 as the transmission of both gossip (heard by four individuals), and knowledge (heard by five individuals) was infrequent in group 2. Characteristics of the Information Source Dominance of Source. As well as exploring whether gossip transmitted more readily than knowledge we were also interested in the impact of source dominance on transmission. In order to explore the influence of dominance we first of all totaled the number of individuals who reported hearing the gossip and knowledge statements directly from the two nondominant sources and the two dominant sources, before going on to explore each of the four sources individually. This revealed that the two dominant sources transmitted the gossip to a total of 10 individuals and the two nondominant sources transmitted the gossip to seven individuals, a difference which was not significant. This pattern of results was mirrored in the performance of the individual sources as little difference was evident between the two male sources (dominant boy told eight individuals; nondominant boy told six individuals), or the two female sources (dominant girl told two individuals; nondominant
Downloaded by [University of Connecticut] at 08:01 11 October 2014
612
The Journal of Genetic Psychology
FIGURE 1. The left-hand side of the figure shows the transmission of gossip and general knowledge in group 1, and the right-hand side shows transmission in group 2. A represents the dominant boy, B represents the nondominant boy, C represents the dominant girl, and D represents the nondominant female. The transmission of gossip is represented with an arrow and the transmission of general knowledge is represented by a circle. Friends are shown with a checked background and nonfriends with a spotted background. M = male recipient, F = female recipient.
McGuigan & Cubillo
613
Downloaded by [University of Connecticut] at 08:01 11 October 2014
girl told one individual), suggesting that dominance level had little impact on gossip transmission. Similarly, the dominance level of the source had little influence on the transmission of knowledge, either across the source type combined (nondominant sources transmitted to three individuals; dominant sources transmitted to five individuals), or at the level of the individual source (dominant boy told one individual; nondominant boy told one individual; dominant girl told four individuals; nondominant girl told two individuals). Sex of Source. The second characteristic of the source that may have influenced the spread of the information was the sex of the source. As the dominance level of the source had little impact on transmission we combined the data from the two male sources and the two female sources to explore the influence of source sex. This revealed that the male sources transmitted the gossip directly to 14 individuals, whereas the two female sources spread the gossip to only three individuals, a difference which was found to be significant (Fisher’s exact test, p = .013). Unfortunately due to the small sample size equivalent statistical comparisons could not be made with respect to the transmission of knowledge, however the trend was toward greater transmission by the female sources (male sources told two individuals; female sources told six individuals). Characteristics of Information Recipient Friend or Nonfriend. As each participant provided a list of friends we could use this information to explore whether information was more readily transmitted to friends or nonfriends. Across the two groups the gossip statements spread to a total of 22 individuals, 13 friends and nine nonfriends, a difference which was not significant. Although the small sample size does not allow for detailed statistical analysis of the individual sources there appeared to be trends in the data, with the dominant male transmitting gossip exclusively to friends (eight friends vs. zero nonfriends), and the nondominant boy transmitting to an equal number of friends and nonfriends (three individuals in each case). The female sources transmitted very little gossip overall but that which did occur was as likely to be a friend (two individuals), or a nonfriend (one individual). In contrast the pattern with respect to the knowledge statements was somewhat different, with the participants indicating that they heard the knowledge statements exclusively from friends rather than nonfriends (eight friends vs. zero nonfriends). Dominance of Recipient. An additional factor that may play a role in the transfer of gossip is the recipient’s peer standing within the group, and their position in relation to the dominance level of the source. In order to explore this relationship we averaged the peer regard scores of the 10 individuals who received the gossip from the two dominant sources, and the seven individuals who received the gossip from the two nondominant sources. This revealed that the peer regard scores of the individuals who received the gossip from the nondominant sources were
614
The Journal of Genetic Psychology
Downloaded by [University of Connecticut] at 08:01 11 October 2014
significantly higher (M peer regard = 86, SD = 33), than those of the individuals who received the gossip from the dominant sources (M peer regard = 68, SD = 23), F(1, 15) = 107, p < .001, μ2 = .097. This was particularly true of the least dominant boy who transmitted the gossip to individuals of a significantly higher social standing (M peer regard score = 89, SD = 31), than the dominant boy (M peer regard score = 68, SD = 24), the dominant girl (M peer regard score = 65, SD = 17), and the nondominant girl (M peer regard score = 18, SD = 0; p < .001 in each case). Sex of Recipient. The final question of interest was whether individuals would be more likely to transmit information to individuals of the same sex as themselves. Across the two groups there was a nonsignificant trend toward the transmission of gossip to same-sex individuals (15 occasions), than opposite-sex individuals (seven occasions). With respect to the individual sources the limited sample size does not allow for statistical analysis. However, the trends in the data appeared to show little variation in the sex of the gossip recipient with the exception of the dominant boy who transmitted to same-sex individuals (six individuals) more frequently than opposite-sex individuals (two individuals). In contrast, the sex of the recipient had very little impact on the transfer of knowledge as the information was transmitted to three same-sex individuals, and five opposite-sex individuals. Discussion The present study used an open diffusion approach to explore the transmission of gossip and general knowledge within two groups of 10- and 11-year-old children. The groups were seeded by two information sources preselected on the basis of their sex, and dominance ranking within their peer group. The results indicated that the type of information was crucial to transmission, with gossip spreading among the children more readily than general knowledge, thus demonstrating a social bias in information transmission earlier than previously witnessed. However, extensive transmission of the gossip occurred in one group only, suggesting that particular characteristics of the sources may have impacted on transmission. One characteristic of the source that seemed particularly influential was the sex of the source, with the male sources transmitting more gossip than the female sources. This finding is somewhat counter to previous literature which tends to report more frequent gossiping by females (e.g., Kless, 1992; Nevo et al., 1993). Of interest is why the male sources transmitted gossip so readily? Conversely why were the female sources more restrained in this respect? The first possibility lies in the content of the gossip itself. The type of information which may be of most interest to males is information that concerns the status of another male, including information that is relevant to resources and physical prowess (McAndrew & Milenkovic, 2002). Arguably the two gossip statements provided to the male sources contained information which was more likely to be about a same sex
Downloaded by [University of Connecticut] at 08:01 11 October 2014
McGuigan & Cubillo
615
other (male teacher), as well as alluding to the gossipee’s resources (Porsche), or physical ability (Olympics). This attribution may have been less likely for the statements provided to the female sources, which could have been perceived as more gender neutral. Perhaps the neutrality of the statements, and the subsequent lack of gender specificity, was less appealing to the female sources. Alternatively, the female sources might have failed to transmit the gossip as the gossip statements did not refer to relationships between individuals. Many studies have shown that a key characteristic of female gossip, particularly female–female gossip, is that it concerns relationships, especially those of family and friends (e.g., Levin & Arluke, 1985b). The gossip provided here lacked such a relational component, potentially downplaying its interest to the female sources. A second explanation for the sex difference witnessed may lie in the different structure of male and female social groups during this age period. Boys of this age tend to have loose, and extensive social groups, whereas girls form tighter social groups, often pair bonds during this period (Lagerspetz, Bj¨orkqvist, & Peltonen, 1988). The loose nature of the male social group may lend itself to a more widespread transmission of gossip, whereas tighter social groups may lead to reduced transmission to a select few. The differences may also be explained by the different ways that aggression is used by each sex, with boys tending to employ more direct overt aggression (e.g., hitting), and girls using more indirect (e.g., negative gossip, back-biting) relational aggression (Bj¨orkqvist, Lagerspetz, & Kaukianen, 1992; Crick, 1996; Crick & Grotpeter, 1995). The greater importance of negative gossip for girls has led to the suggestion that gossip is key to maintaining status in female groups (Eckert, 1990; Leaper & Holliday, 1995), with the result that young females may be more attuned to the truthfulness of gossip than young males (Kuttler et al., 2002). This might have resulted in the female sources being more skeptical about the truthfulness of the gossip, and subsequently more reluctant to transmit the information to other individuals. As well as exploring the role of source characteristics we were also interested in who the information was transmitted to, as well as the relationship between the gossiper and the gossip recipient. The first question of interest was whether the information was as likely to be transmitted to a friend or a nonfriend. Previous research has been somewhat contradictory in this respect, with some studies suggesting that gossip is more likely to occur between friends (Leaper & Holliday, 1995), and other studies suggesting that children are as likely to gossip with friends as they are acquaintances (Mettetal, 1983). The present study suggests that there may be merit in both claims depending on the function of gossip in each case. Looking across the two groups we see a nonsignificant trend toward transmission of gossip to friends. However, when looking at the sources individually the dominant boy transmitted exclusively to friends whereas the nondominant boy, whereas both of the female sources were as likely to transmit to friends as they were nonfriends. This finding supports the suggestion that the transmission of gossip may have been serving different functions in each case. One key feature of gossip is
Downloaded by [University of Connecticut] at 08:01 11 October 2014
616
The Journal of Genetic Psychology
that it can serve to strengthen the social bonds with a group (Ben-Ze’ev, 1994). The dominant boy is likely to have an extensive social network, and by passing the gossip to members of this network the group bonds may have been strengthened. In contrast, the position may be somewhat different for the nondominant boy, who is likely to have a smaller social network, and may have been attempting to use the gossip as a tool to acquire friendships (McAndrew et al., 2007). In addition to using gossip as a tool to acquire new friends it appeared that the nondominant boy might have been attempting to use gossip strategically in order to enhance his social status within the group. Barkow (1992) suggested that from the preschool years onward individuals generally attempt to improve our social standing with our group. Social dominance is extremely desirable as dominant individuals are viewed as more attractive social partners (LaFreniere & Charlesworth, 1983), are looked at (Hawley & Little, 1999; Vaughn & Waters, 1981), and imitated more by peers (Abramovitch & Grusec, 1978), and have greater success in controlling resources (Hawley, 2002). One mechanism an individual could use in an attempt to enhance social status is to impart novel social information to a more dominant, sought after, member of the social group (McAndrew et al., 2007). This claim is supported by the tendency of the low-ranking boy to spread gossip to individuals with a higher peer regard score than did the dominant boy. Overall, the present results provide the first empirical evidence of a social bias the transmission of information in childhood. However, the results appear to suggest that the transmission of social information is multifaceted and highly dependent on many different factors including the characteristics of the gossiper, characteristics of the gossip recipient, and the relationship between the two in terms of friendship and relative social standing. This complexity could be explained from two, potentially intimately related, perspectives. One possible explanation may lie at the level of the child’s social world, in particular from the perspective of social learning. More specifically, within the present study social learning may be evident in relation to the content of the information as preferential attention to the gossip statements may have provided the children with potentially useful information about successful, salient individuals (i.e., teachers) in their social world. Similarly, a social learning explanation could also account for the pattern of transmission witnessed within the peer group itself. Abramovitch and Grusec (1978) showed that children in the present age range who are rated as most dominant within their group are looked at and imitated more frequently than nondominant individuals. A potential explanation for this differential treatment is that dominant individuals are seen as prized sources of knowledge, leading other individuals in the group to monitor their behavior closely so that they can learn appropriate behaviors from them. If we combine this social learning tendency with the finding that boys participate in larger social groups than girls, then it would appear that boys have greater opportunity to take an interest in and attend to the behavior of other individuals. This increased level of opportunity may explain the greater levels of transmission by the dominant male source in the present study.
McGuigan & Cubillo
617
Downloaded by [University of Connecticut] at 08:01 11 October 2014
Intriguingly the patterns of performance witnessed also fit well with the theory that the way in which information is transmitted within groups is underpinned by evolved psychological mechanisms (e.g., Richerson & Boyd, 2005) which bias the transmission of information both at the level of the content of the information (content-based bias) and at the level of the information provider (model-based bias). These intriguing findings, although preliminary, pave the way for more detailed explorations of the relationship between source characteristics and the content of the information to be transmitted. AUTHOR NOTES Nicola McGuigan is a lecturer in the department of psychology, School of Life Sciences at Heriot Watt University in Scotland. Her research interests lie in social learning and more recently in prosocial behavior. Marcus Cubillo is a researcher in the department of psychology, School of Life Sciences at Heriot Watt University. His research interests include information transmission within groups. REFERENCES Abramovitch, R., & Grusec, J. E. (1978). Peer imitation in a natural setting. Child Development, 49, 60–65. Barkow, J. H. (1992). Beneath new culture is old psychology: Gossip and social stratification. In J. H. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 627–637). New York, NY: Oxford University Press. Ben-Ze’ev, A. (1994). The vindication of gossip. In R. F. Goodman & A. Ben-Ze’ev (Eds.), Good gossip (pp. 11–24). Lawrence, KS: University Press of Kansas. Bj¨orkqvist, K., Lagerspetz, K. M. J., & Kaukianen, A. (1992). Do girls manipulate and boys fight? Developmental trends in regard to direct and indirect aggression. Aggressive Behavior, 18, 117–127. Byrne, R. W., & Whiten, A. (1988). Machiavellian intelligence-social expertise and the evolution of intellect in monkeys, apes and humans. Oxford, England: Claredon. Crick, N. R. (1996). The role of overt aggression, relational aggression, and prosocial behavior in the prediction of children’s future social adjustment. Child Development, 67, 2317–2327. Crick, N. R., & Grotpeter, J. K. (1995). Relational aggression, gender, and socialpsychological adjustment. Child Development, 66, 710–722. De Backer, C., Neliseen, M., Vyncke, P., Braeckman, J., & McAndrew, F. (2007). Celebrities: From teachers to friends. Human Nature, 18, 334–354. Dunbar, R. I. M. (1993). Coevolution of neocortical size, group size and language. Behavioral and Brain Sciences, 16, 681–735. Dunbar, R. I. M. (1996). Grooming, gossip and the evolution of language. London, England: Faber and Faber. Dunbar, R. I. M. (1998). The social brain hypothesis. Evolutionary Anthropology, 6, 178–190. Dunbar, R. I. M. (2003). The social brain: Mind, language, and society in evolutionary perspective. Annual Review of Anthropology, 32, 163–181.
Downloaded by [University of Connecticut] at 08:01 11 October 2014
618
The Journal of Genetic Psychology
Dunbar, R. I. M. (2004). Gossip in evolutionary perspective. Review of General Psychology, 8, 100–110. Eckert, P. (1990). Cooperative competition in adolescent “girl talk.” Discourse Processes, 13, 91–122. Eder, D., & Enke, J. L. (1991). The structure of gossip: Opportunities and constraints on collective expression among adolescents. American Sociological Review, 56, 494–508. Fine, G. A., & Rosnow, R. L. (1978). Gossip, gossipers, gossiping. Personality and Social Psychology Bulletin, 4, 161–168. Flynn, E. (2008). Investigating children as cultural magnets: do young children transmit redundant information along diffusion chains? Philosophical Transmissions of The Royal Society, 363, 3541–3551. Flynn, E., & Whiten, A. (2008) Cultural transmission of tool use in young children: A diffusion chain study. Social Development, 17, 699–718. Hawley, P. H. (2002). Social dominance and prosocial and coercive strategies of resource control in preschoolers. International Journal of Behavioral Development, 26, 167–176. Hawley, P. H., & Little, T. D. (1999). Winning some and losing some: A social relations approach to social dominance in toddlers. Merrill-Palmer Quarterly, 45, 185–214. Henrich, J., & McElreath, R. (2007). Dual-inheritance theory: The evolution of human cultural capacities and cultural evolution. In R. Dunbar & L. Barrett (Eds.), Oxford handbook of evolutionary psychology (pp. 555–570). Oxford, England: Oxford University Press. Horner, V., Whiten, A., Flynn, E., & de Waal, F. B. M. (2006). Faithful replication of foraging techniques along cultural transmission chains by chimpanzees and children. National Academic Sciences, 103, 13878–13883. Kless, S. J. (1992). The attainment of peer status: Gender and power relationships in the elementary school. Sociological Studies of Child Development, 5, 115–148. Kuttler, A. F., Parker, J. G., & La Greca, A. M. (2002). Developmental and gender differences in preadolescents’ judgments of the veracity of gossip. Merrill-Palmer Quarterly, 48, 105–132. LaFreniere, P. J., & Charlesworth, W. R. (1987). Effects of friendship and dominance status on preschooler’s resource utilization in a cooperative/competitive paradigm. International Journal of Behavioral Development, 10, 345–358. Lagerspetz, K. M. J., Bj¨orkqvist, K., & Peltonen, T. (1988). Is indirect aggression typical of females? Gender differences in aggressiveness in 11- to 12-year-old children. Aggressive Behavior, 14, 403–414. Leaper, C., & Holliday, H. (1995). Gossip in a same-gender and cross gender friends’ conversations. Personal Relationships, 2, 237–246. Levin, J., & Arluke, A. (1985a). Gossip: The inside scoop. New York, NY: Plenum. Levin, J., & Arluke, A. (1985b). An exploratory analysis of sex differences in gossip. Sex Roles, 12, 281–286. McAndrew, F. T., Bell, E. K., & Garcia, C. M. (2007). Who do we tell and whom do we tell on? Gossip as a strategy for status enhancement. Journal of Applied Social Psychology, 37, 1562–1577. McAndrew, F. T., & Milenkovic, M. A. (2002). Of tabloids and family secrets: The evolutionary psychology of gossip. Journal of Applied Social Psychology, 32, 1064–1082. McGuigan, N., & Graham, M. (2010). Cultural transmission of irrelevant tool actions in diffusion chains of 3- and 5-year-old children. European Journal of Developmental Psychology, 7, 561–577. Mesoudi, A., & Whiten, A. (2008). Establishing an experimental science of culture: Animal social diffusion experiments. Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences, 363, 3489–3502.
Downloaded by [University of Connecticut] at 08:01 11 October 2014
McGuigan & Cubillo
619
Mesoudi, A., Whiten, A., & Dunbar, R. I. M. (2006). A bias for social information in human cultural transmission. British Journal of Psychology, 97, 405–423. Mettetal, G. (1983). Fantasy, gossip, and self-disclosure: Children’s conversations with friends. Communication Yearbook, 7, 717–735. Nevo, O., Nevo, B., & Derech-Zehavi, A. (1993). The development of the tendency to gossip. questionnaire: Construct and concurrent validity for a sample of Israeli college students. Educational and Psychological Measurement, 53, 973–981. Richerson, P. J., & Boyd, R. (2005). Not by genes alone. Chicago, IL: University of Chicago Press. Strayer, F. F., & Trudel, M. (1984). Developmental changes in the nature and function of social dominance among young children. Ethology and Sociobiology, 5, 279–295. Vaughn, B. E., & Waters, E. (1981). Attention structure, sociometric status, and dominance: Interrelations, behavioral correlates, and relationships to social competence. Developmental Psychology, 17, 275–288. Wert, S. R., & Salovey, P. (2004). A social comparison account of gossip. Review of General Psychology, 8, 122–137. Whiten, A. (1999). The evolution of deep social mind in humans. In M. C. Corballis & S. E. G. Lea (Eds.), The descent of mind: Psychological perspectives on hominid evolution (pp. 173–193). Oxford, England: Oxford University Press. Whiten, A., & Byrne, R. W. (Eds.). (1997). Machiavellian intelligence II: Extensions and evaluations. New York, NY: Cambridge University Press. Whiten, A., & Flynn, E. G. (2010). The transmission and evolution of experimental microcultures in groups of young children. Developmental Psychology, 46, 1694–1709. Whiten, A., McGuigan, N., Marshall-Pescini, S., & Hopper, L. (2009). Emulation, imitation, over-imitation and the scope of culture for child and chimpanzee. Philosophical Transactions of the Royal Society B, 364, 2117–2428. Whiten, A., & Mesoudi, A. (2008). An experimental science of culture: Animal social diffusion experiments. Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences, 363, 3477–3488.
Original manuscript received July 16, 2012 Final version accepted November 13, 2012
